This dissertation studies the feature-based image comparison method and its application in Wireless Visual Sensor Networks. Wireless Visual Sensor Networks (WVSNs), formed by a large number of low-cost, small-size visual sensor nodes, represent a new trend in surveillance and monitoring practices. Although each single sensor has very limited capability in sensing, processing and transmission, by working together they can achieve various high level tasks. Sensor collaboration is essential to WVSNs and normally performed among sensors having similar measurements, which are called neighbor sensors. The directional sensing characteristics of imagers and the presence of visual occlusion present unique challenges to neighborhood formation, as geographically-close neighbors might not monitor similar scenes. Besides, the energy resource on the WVSNs is also very tight, with wireless communication and complicated computation consuming most energy in WVSNs. Therefore the feature-based image comparison method has been proposed, which directly compares the captured image from each visual sensor in an economical way in terms of both the computational cost and the transmission overhead. The feature-based image comparison method compares different images and aims to find similar image pairs using a set of local features from each image. The image feature is a numerical representation of the raw image and can be more compact in terms of the data volume than the raw image. The feature-based image comparison contains three steps: feature detection, descriptor calculation and feature comparison. For the step of feature detection, the dissertation proposes two computationally efficient corner detectors. The first detector is based on the Discrete Wavelet Transform that provides multi-scale corner point detection and the scale selection is achieved efficiently through a Gaussian convolution approach. The second detector is based on a linear unmixing model, which treats a corner point as the intersection of two or three “line” bases in a 3 by 3 region. The line bases are extracted through a constrained Nonnegative Matrix Factorization (NMF) approach and the corner detection is accomplished through counting the number of contributing bases in the linear mixture. For the step of descriptor calculation, the dissertation proposes an effective dimensionality reduction algorithm for the high dimensional Scale Invariant Feature Transform (SIFT) descriptors. A set of 40 SIFT descriptor bases are extracted through constrained NMF from a large training set and all SIFT descriptors are then projected onto the space spanned by these bases, achieving dimensionality reduction. The efficiency of the proposed corner detectors have been proven through theoretical analysis. In addition, the effectiveness of the proposed corner detectors and the dimensionality reduction approach has been validated through extensive comparison with several state-of-the-art feature detector/descriptor combinations.

Eyewitness recognition is acknowledged to be prone to error but there is less understanding of difficulty in discriminating unfamiliar faces. This thesis examined the effects of face perception on identification of facial composites, and on unfamiliar face image comparison. Facial composites depict face memories by reconstructing features and configurations to form a likeness. They are generally reconstructed from an unfamiliar face memory, and will be unavoidably flawed. Identification will require perception of any accurate features, by someone who is familiar with the suspect and performance is typically poor. In typical face perception, face images are processed efficiently as complete units of information. Chapter 2 explored the possibility that holistic processing of inaccurate composite configurations will impair identification of individual features. Composites were split below the eyes and misaligned to impair holistic analysis (cf. Young, Hellawell, & Jay, 1987); identification was significantly enhanced, indicating that perceptual expertise with inaccurate configurations exerts powerful effects that can be reduced by enabling featural analysis. Facial composite recognition is difficult, which means that perception and judgement will be influence by an affective recognition bias: smiles enhance perceived familiarity, while negative expressions produce the opposite effect. In applied use, facial composites are generally produced from unpleasant memories and will convey negative expression; affective bias will, therefore, be important for facial composite recognition. Chapter 3 explored the effect of positive expression on composite identification: composite expressions were enhanced, and positive affect significantly increased identification. Affective quality rather than expression strength mediated the effect, with subtle manipulations being very effective. Facial image comparison (FIC) involves discrimination of two or more face images. Accuracy in unfamiliar face matching is typically in the region of 70%, and as discrimination is difficult, may be influenced by affective bias. Chapter 4 explored the smiling face effect in unfamiliar face matching. When multiple items were compared, positive affect did not enhance performance and false positive identification increased. With a delayed matching procedure, identification was not enhanced but in contrast to face recognition and simultaneous matching, positive affect improved rejection of foil images. Distinctive faces are easier to discriminate. Chapter 5 evaluated a systematic caricature transformation as a means to increase distinctiveness and enhance discrimination of unfamiliar faces. Identification of matching face images did not improve, but successful rejection of non-matching items was significantly enhanced. Chapter 6 used face matching to explore the basis of own race bias in face perception. Other race faces were manipulated to show own race facial variation, and own race faces to show African American facial variation. When multiple face images were matched simultaneously, the transformation impaired performance for all of the images; but when images were individually matched, the transformation improved perception of other race faces and discrimination of own race faces declined. Transformation of Japanese faces to show own race dimensions produced the same pattern of effects but failed to reach significance. The results provide support for both perceptual expertise and featural processing theories of own race bias. Results are interpreted with reference to face perception theories; implications for application and future study are discussed.

Automobilių apsauga visada buvo, yra ir bus aktuali. Esamos apsaugos priemonės nesunkiai apeinamos ir nesudaro didelių kliūčių, norint transporto priemonę pasisavinti. Reikalinga naujoviška vairuotojo autentifikavimo sistema, galinti sumažinti šią, pasisavinimo tikimybę. Tam tinka biometrija – asmens bruožų ar charakteristikų statistinė analizė ir nustatymas. Juos nustačius, gali būti naudojami asmens autentifikacijai. Tai daroma lyginant su anksčiau užfiksuotu šablonu. Naudojama šioms autentifikacijoms: kompiuteriuose tinklų, praėjimo kontrolės, bankomatuose, apsipirkimams kreditinėmis kortelėmis, mobiliuosiuose telefonuose, delniniuose kompiuteriuose, medicininių įrašų valdymui, nuotoliniame mokymesi ir t.t. Pirštų kraujagyslės – nauja biometrinės autentifikacijos forma. Kadangi kraujagyslės yra po oda, jos yra nematomos apšviečiant natūralia šviesa. Tačiau jos puikiai matomos apšvietus artima infraraudoniems spinduliams šviesa. Šiame tyrime buvo sukurtas ir ištirtas biometrinės autentifikacijos automobiliams modelis. Ištirtas trijų atvaizdų apdorojimo algoritmų tinkamums, nustatant kuris jų mažiausia jautrus raiškos ir triukšmo pasikeitimams. Jį ir siūlome naudoti šiame autentifikacijos modelyje. / Automobile security is always relevant. Existing security systems could be easily bypassed and are not a big barrier for thieves. So there is a necessity for a innovative authentication system. Biometrics consist of methods for uniquely recognizing humans based upon one or more intrinsic physical or behavioral traits. In computer science, in particular, biometrics is used as a form of identity access management and access control. It could be also used for automobile security. Finger vein comparison is a relatively new form of biometrics, which could also be used for authentication. Images of finger veins could be obtained illuminating fingers with near-infrared light. Those images are very unique and different for every person and every finger. In this research biometric authentication model for automobiles is developed and researched. Three image recognition algorythms were tested in the experiment, determining which one is less sensitive to resolution changes and noise. The results were given, proposing use of one algorythm, in our model of biometric authentication system for automobiles.

L'objectif de cette thèse est de proposer une solution au problème de la constance des couleurs entre les images d'une même scène acquises selon un même point de vue ou selon différents points de vue. Ce problème constitue un défi majeur en vision par ordinateur car d'un point de vue à l'autre, on peut être confronté à des variations des conditions d'éclairage (spectre de l'éclairage, intensité de l'éclairage) et des conditions de prise de vue (point de vue, type de caméra, paramètres d'acquisition tels que focus, exposition, balance des blancs, etc.). Ces variations induisent alors des différences d'apparence couleur entre les images acquises qui touchent soit sur l'ensemble de la scène observée soit sur une partie de celle-ci. Dans cette thèse, nous proposons une solution à ce problème qui permet de modéliser puis de compenser, de corriger, ces variations de couleur à partir d'une méthode basée sur quatre étapes : (1) calcul des correspondances géométriques à partir de points d'intérêt (SIFT et MESR) ; (2) calculs des correspondances couleurs à partir d'une approche locale; (3) modélisation de ces correspondances par une méthode de type RANSAC; (4) compensation des différences de couleur par une méthode polynomiale à partir de chacun des canaux couleur, puis par une méthode d'approximation linéaire conjuguée à une méthode d'estimation de l'illuminant de type CAT afin de tenir compte des intercorrélations entre canaux couleur et des changements couleur dus à l'illuminant. Cette solution est comparée aux autres approches de l'état de l'art. Afin d'évaluer quantitativement et qualitativement la pertinence, la performance et la robustesse de cette solution, nous proposons deux jeux d'images spécialement conçus à cet effet. Les résultats de différentes expérimentations que nous avons menées prouvent que la solution que nous proposons est plus performante que toutes les autres solutions proposées jusqu'alors / The objective of this thesis is to propose a solution to the problem of color consistency between images originate from the same scene irrespective of acquisition conditions. Therefore, we present a new color mapping framework that is able to compensate color differences and achieve color consistency between views of the same scene. Our proposed, new framework works in two phases. In the first phase, we propose a new method that can robustly collect color correspondences from the neighborhood of sparse feature correspondences, despite the low accuracy of feature correspondences. In the second phase, from these color correspondences, we introduce a new, two-step, robust estimation of the color mapping model: first, nonlinear channel-wise estimation; second, linear cross-channel estimation. For experimental assessment, we propose two new image datasets: one with ground truth for quantitative assessment; another, without the ground truth for qualitative assessment. We have demonstrated a series of experiments in order to investigate the robustness of our proposed framework as well as its comparison with the state of the art. We have also provided brief overview, sample results, and future perspectives of various applications of color mapping. In experimental results, we have demonstrated that, unlike many methods of the state of the art, our proposed color mapping is robust to changes of: illumination spectrum, illumination intensity, imaging devices (sensor, optic), imaging device settings (exposure, white balance), viewing conditions (viewing angle, viewing distance)

L’essor des nouvelles techniques de traitement en radiothérapie externe a rendu nécessaire la mise en place de nouveaux contrôles qualité (CQ). Il est en effet capital de s’assurer que la dose délivrée au patient est bien conforme à celle qui lui a été prescrite. Les détecteurs EPID, initialement utilisés pour contrôler l’exactitude du positionnement du patient, sont de plus en plus employés pour vérifier la conformité du traitement. L’image qu’ils enregistrent au cours d’une irradiation peut par exemple être comparée à une image de référence, qui correspond à ce qui aurait été mesuré si le traitement s’était déroulé dans les conditions de sa planification. Le succès de ce CQ repose (1) sur la précision avec laquelle on peut prédire l’image EPID (ou portale) de référence et (2) sur les performances de l’outil de comparaison d’image utilisé. Nous avons étudié au cours de cette thèse ces deux points clés. Nous nous sommes tout d’abord tournés vers une méthode de prédiction d’images EPID haute résolution basée sur le couplage de simulations Monte-Carlo (MC) et de la technique de débruitage DGPLM. Pour la mettre en œuvre, nous avons modélisé un accélérateur linéaire d’électrons à usage médical (linac) dans le code MC PENELOPE et optimisé les paramètres de sa source d’électrons primaires pour des calculs de dose dans l’eau. L’analyse d’un fichier d’espace des phases (PSF) de 71 Go stocké sous le cône égalisateur nous a ensuite permis de développer un modèle de sources virtuelles (MSV) représenté par des histogrammes corrélés (environ 200 Mo). Ce nouveau MSV, plus compact que le PSF, est tout aussi précis pour les calculs de dose dans l’eau si son maillage est déterminé selon une méthode adaptative. La modélisation du détecteur EPID dans PENELOPE suggère que les hypothèses faites sur les propriétés de la tâche focale du linac sont trop simplistes et doivent être reconsidérées. L’évaluation du MSV pour la prédiction d’images EPID haute résolution a quant à elle conduit à d’excellents résultats. Une fois la chaine de prédiction de l’image portale de référence validée, nous l’avons utilisée pour détecter des irrégularités dans les traitements de RCMI. Dans une étude préliminaire, nous avons volontairement introduit des erreurs de traitement dans le calcul d’images EPID (dérive du faisceau d’irradiation, modification de la morphologie ou de la position du patient). Le γ-index traditionnellement utilisé en routine s’est avéré moins performant que le χ-index pour les détecter. Une étude plus approfondie aura pour objet de déterminer des seuils de détection d’erreurs en fonction de leur nature et d’éprouver les performances d’autres tests. / Advanced techniques used in radiotherapy for cancer treatment, such as Intensity-Modulated Radiation Therapy (IMRT), require extensive verification procedures to ensure the correct dose delivery. Electronic Portal Imaging Devices (EPIDs) are widely used for quality assurance in radiotherapy, and also for dosimetric verifications. For this latter application, the images obtained during the treatment session can be compared to a pre-calculated reference image in order to highlight dose delivery errors. The quality control performance depends (1) on the accuracy of the pre-calculated reference image (2) on the ability of the tool used to compare images to detect errors. These two key points were studied during this PhD work. We chose to use a Monte Carlo (MC)-based method developed in the laboratory and based on the DPGLM (Dirichlet process generalized linear model) denoising technique to predict high-resolution reference images. A model of the studied linear accelerator (linac Synergy, Elekta, Crawley, UK) was first developed using the PENELOPE MC codes, and then commissioned using measurements acquired in the Hôpital Nord of Marseille. A 71 Go phase space file (PSF) stored under the flattening filter was then analyzed to build a new kind of virtual source model based on correlated histograms (200 Mo). This new and compact VSM is as much accurate as the PSF to calculate dose distributions in water if histogram sampling is based on adaptive method. The associated EPID modelling in PENELOPE suggests that hypothesis about linac primary source were too simple and should be reconsidered. The use of the VSM to predict high-resolution portal images however led to excellent results. The VSM associated to the linac and EPID MC models were used to detect errors in IMRT treatment plans. A preliminary study was conducted introducing on purpose treatment errors in portal image calculations (primary source parameters, phantom position and morphology changes). The γ-index commonly used in clinical routine appears to be less effective than the χ-index. A future in-depth study will be dedicated to determine error detection threshold according to their nature and to evaluate other comparison test robustness. The developed portal image prediction method associated to robust analysis tools will then constitute an adapted way to assure treatment quality control.

In March 2019 the 5th direct presidential elections were held in Slovakia. In the first round there were 13 candidates running for the office. The winner after the second round held on March 30th was Zuzana Čaputová. With 58,4 percent of all valid votes she became the first female president of The Slovak republic on June 15. This thesis goes into the comparison of Zuzana Čaputová's media image since being elected on the most visited news web sites in The Czech republic and Slovakia. The goal of the thesis is to compare the quantity and content of information published in both countries and to track possible changes in the media narrative and what has caused them. Using the quantitative method of content analysis and qualitative method of frame analysis of relevant articles published on news websites Novinky.cz and Aktuality.sk I want to monitor the decline, stagnation or rise of media interest in the persona of Zuzana Čaputová in the selected time period. This will help to describe the media image of the president on the Czech and Slovak media scene. The source articles are attached to the thesis.